ABSTRACT Like other cancers, T-cell acute lymphoblastic leukemia (T-ALL) arises from the accumulation of genetic abnormalities that impair function of immature T-cell progenitors. DNMT3A, which encodes a de novo DNA methyltransferase enzyme that catalyzes the establishment of new DNA methylation marks on the genome, is recurrently mutated in 10-18% of adult T-ALL cases and confers a poor clinical prognosis. We recently showed using genetic mouse models that Dnmt3a acts as a T-cell tumor suppressor. Introduction of an activating Notch1 mutation into a Dnmt3a loss-of-function genetic background (a common genetic combination in patients) generated a lethal T-ALL with half the latency period compared to T-ALL with wild-type Dnmt3a. Dnmt3a-mutant T-ALL blasts are resistant to cell death, both in vivo and under different stress conditions. This suggests a major biological function of DNMT3A mutations in T-ALL cells is to make them ?harder to kill?, and the inferior clinical outcomes of these patients may be due to resistance to standard chemotherapy regimens. We have now uncovered a potential mechanism to explain this which is conserved in both Dnmt3a-mutant mouse T-ALL models and primary human DNMT3A-mutant T-ALL patient samples. DNMT3A-mutant T-ALL cells are hypersensitive to cytokines such as IL-6 and IL-7, which results in elevated JAK/STAT signaling triggering a pro-survival gene expression program. Specifically, we hypothesize that DNMT3A-mutant T- ALL cells are resistant to apoptosis due to pSTAT5-dependent upregulation of BCL-xL. The goals of this proposal are to understand the molecular mechanisms driving these phenotypes and exploit them for novel therapeutic interventions. We will test this hypothesis with the following Specific Aims; ? Determine the role of STAT5 and BCL-xL in the pathogenesis of DNMT3A-mutant T-ALL ? Define the role of JAK/STAT signaling in chemoresistance of DNMT3A-mutant T-ALL. ? Identify molecular mechanisms underpinning the functional phenotypes of Dnmt3a-mutant T-ALL. We will leverage preliminary findings to interrogate this hypothesis using a complementary combination of genetic mouse models, human patient-derived xenografts, and CRSIPR/Cas9 genome engineering. In Aim 1, we will evaluate the importance of STAT5 and BCL-xL for the development and maintenance of DNMT3A- mutant T-ALL (both mouse and human) using state-of-the-art genetic tools. In Aim 2, we will use genomic assays to understand how DNMT3A-mutant T-ALL cells are resistant to chemotherapy, and if JAK/STAT inhibition can resensitize these cells to chemotherapeutic agents in vivo. In Aim 3, we will determine the importance of DNA methylation at enhancers in the generation of pathogenic gene expression programs for this T-ALL subtype, and evaluate the role of SHP-1 in conferring cytokine hypersentivity to DNMT3A-mutant T- ALL cells. We will use the results of this project to inform design of rationally-targeted precision medicine strategies for the treatment of T-ALL patients based on their underlying genetics.